CC Chemokine Receptor-3 Undergoes Prolonged Ligand-induced Internalization*

CC chemokine receptor-3 (CCR-3) is a major receptor involved in regulating eosinophil trafficking; therefore, elucidation of ligand-induced CCR-3 events has important implications in understanding the biological and pathological properties of eosinophils. Previous studies have demonstrated that unique receptor events occur in different cell types supporting investigation of CCR-3-mediated events in eosinophilic cells. We now report biochemical characterization of CCR-3 internalization following exposure of eosinophils to CCR-3 ligands. Treatment of freshly isolated human eosinophils with CCR-3 ligands resulted in marked and differential internalization of CCR-3 in a dose-dependent manner. Exposure to 100 ng/ml eotaxin reduced surface expression to 43, 43, and 76% at 15 min, 1 h, and 3 h, respectively. RANTES (reduced on activation T cell expressed and secreted) treatment induced more significant and prolonged internalization of CCR-3 than eotaxin; following 100 ng/ml of RANTES, 29, 24, and 47% of the receptor was expressed at 15 min, 3 h, and 18 h, respectively. Confocal microscopy demonstrated that receptor modulation involved receptor internalization by an endocytic pathway shared with the transferrin receptor. Receptor internalization was accompanied by partial degradation of CCR-3, and reexpression of CCR-3 was dependent in part upon de novo protein synthesis. Internalization was not blocked by pretreatment of eosinophils with pertussis toxin. Furthermore, staurosporine did not inhibit internalization although it blocked phorbol 12-myristate 13-acetate-induced CCR-3 down-modulation. These results demonstrate that CCR-3 ligands induce differential receptor internalization that is not dependent upon Gi-protein coupling, calcium transients, or protein kinase C.

Eosinophils are bone marrow-derived granulocytes associated with numerous diseases including allergic disorders, parasitic infections, and malignancies (1)(2)(3). Elucidating the processes that regulate eosinophil tissue accumulation is fundamental, since these cells markedly increase in tissue locations and cause potent proinflammatory effects in numerous diseases.
Recent studies have focused on the orchestration of eosinophil tissue accumulation by chemokines, especially eotaxin, the most selective eosinophil chemoattractant identified to date (4,5). Chemokines are grouped into the CXC, CC, C, and CX 3 C subfamilies on the basis of the arrangement of the conserved cysteine residues (6 -8). The CXC and CC groups, in contrast to the C and CX 3 C groups, contain many members and have been studied in detail. The specific effects of chemokines are mediated by a family of seven-transmembrane spanning G-proteincoupled receptors (GPCR) 1 (9,10). To date, 17 chemokine receptors have been described: receptors for the CXC chemokines CXCR-1 through -5; the CC chemokines CCR-1 through -10; the receptor for CX 3 C chemokine CX 3 CR-1; and the C chemokine XCR-1. Most CXC chemokines are active on neutrophils, while CC chemokines have variable potencies for monocytes, lymphocytes, eosinophils, and basophils. In addition to mediating leukocyte chemoattraction and activation, selected chemokine receptors (e.g. CXCR-4, CCR-2, CCR-3, and CCR-5) serve as co-receptors for the entry of human immunodeficiency virus type 1 into cells (11).
The major chemokine receptor operational in eosinophils is CCR-3. This receptor appears to play a central role in allergic responses, since it is not only expressed on eosinophils but also on basophils and Th2 lymphocytes, other cells central in allergic responses (12, 14 -17). CCR-3 binds multiple ligands including eotaxin, RANTES, MCP-2, MCP-3, and MCP-4. Of these chemokines, only eotaxin signals exclusively through CCR-3. Eotaxin has been shown to be responsible for eosinophil trafficking during base-line and inflammatory processes (18 -23).
Following ligand binding to GPCRs, cellular responses are rapidly attenuated. This may be particularly important in eosinophils, since treatment of eosinophils with eotaxin, a chemokine specific for CCR-3, for only 10 min in vitro results in chemokine unresponsiveness for at least 8 h when the cells are adoptively transferred in vivo (21). A variety of mechanisms may be responsible for signal attenuation including receptor desensitization, endocytosis, and down-regulation (24). Each receptor and each cell type utilize unique mechanisms indicating the importance of elucidating individual receptor events in the appropriate cell type (25). It is therefore important to analyze CCR-3-mediated events in eosinophilic cells rather than CCR-3-transfected heterologous cells. We now report biochemical analysis of CCR-3 internalization following exposure of human eosinophils to CCR-3 ligands.

EXPERIMENTAL PROCEDURES
Cell Culture-The AML14.3D10 cell line (kindly provided by C. Paul and M. Baumann, Dayton Veterans Affairs Medical Center, Dayton, OH) was grown in RPMI 1640 (Life Technologies, Inc.) containing 10% * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
¶ To whom correspondence should be addressed: Division of Pulmonary Medicine, Allergy and Clinical Immunology, Department of Pediatrics, Children's Hospital Medical Center, 3333 Burnet Ave., Cincinnati, OH 45229. Tel.: 513-636-7177; Fax: 513-636-3310; E-mail: rothm0@chmcc.org. fetal calf serum (Life Technologies), 50 M 2-mercaptoethanol (Sigma), 0.1 mM nonessential amino acids (Life Technologies), 1 mM sodium pyruvate (Sigma), and penicillin-streptomycin (Life Technologies). Initially, the pH of the media was titrated to 7.8, but after it was established that CCR-3 expression does not depend upon the pH of the media (data not shown) this practice was discontinued. A stock of 50 mM butyric acid (Sigma) in phosphate-buffered saline (PBS) was prepared and stored at 4°C prior to use. IL-5 (R&D, Minneapolis, MN) was prepared as a stock of 100 g/ml and stored at Ϫ80°C prior to use. Induction experiments were performed by growing cells in six-well plates on glass coverslips, starting with a concentration of ϳ10 5 cells in 3 ml/well. Cells were induced with 0.5 mM butyric acid. 2 days later, 10 ng/ml IL-5 was added, and experiments were performed on day 7. Medium was not replenished during the induction period. These cells will be referred to as dAML.
Eosinophil Purification-Eosinophils were purified from healthy or mildly atopic volunteers by negative immunomagnetic selection based on the method of Hansel et al. (26). Briefly, granulocytes were isolated from heparin-anticoagulated whole blood by dextran sedimentation, Percoll centrifugation, and hypotonic lysis of red blood cells. Cells were resuspended in Hanks' buffered salt solution (Life Technologies, Inc.) with 2% fetal calf serum and incubated with 0.75 l/10 6 cells of anti-CD16-conjugated microbeads (MACS; Miltenyi Biotech Inc., Sunnyvale, CA) for 30 min at 4°C. The cell suspension was then applied onto a CS MACS column, and negative populations were collected through a magnetic field. The isolates routinely contained Ͼ95% eosinophils with viability of Ͼ95% as assessed by trypan blue exclusion. For internalization experiments, freshly isolated eosinophils were plated at 0.5 ϫ 10 6 /ml in the same medium used for AML cells but supplemented with 50 pM IL-5. Cells were cultured (37°C, 5% CO 2 ) for a maximum of 22 h, and viability was Ͼ95%.
Intracellular [Ca 2ϩ ] Measurement-Cells (2 ϫ 10 6 /ml) were loaded with 5 M Fura-2/AM (Molecular Probes, Inc., Eugene, OR) in Hanks' buffered salt solution with 1% fetal calf serum for 60 min at 37°C in the dark. After two washes in flux buffer (145 mM NaCl, 4 mM KCl, 1 mM NaHPO 4 , 0.8 mM MgCl 2 , 1.8 mM CaCl 2 , 25 mM Hepes, and 22 mM glucose), cells were resuspended at 2 ϫ 10 6 cells/ml and maintained on ice. Cells (2 ml) were prewarmed to 37°C and stimulated in a cuvette with a continuously stirring magnetic bar using a RatioMaster fluorimeter (Photon Technology, Inc., South Brunswick, NJ). Data were recorded as the relative ratio of fluorescence emitted at 510 nm after excitation at 340 and 380 nm (y axis) over time (x axis).
Flow Cytometry-Cells (5 ϫ 10 5 ) were washed with FACS buffer (2% bovine serum albumin, 0.1% sodium azide in PBS) and incubated with 0.5 g of anti-human CCR-3 antibody (clone 7B11, kindly provided by Dr. Paul Ponath, Leukosite, Cambridge, MA), 0.5 g of anti-CD18 antibody (clone TS1/18, ATCC), or the mouse isotype-matched control IgG2a or IgG1, respectively (Pharmingen, San Diego, CA) for 30 min at 4°C. After two washes in FACS buffer, cells were incubated with 0.5 g of fluorescein isothiocyanate-conjugated isotype-specific secondary antibody (Pharmingen) for 30 min at 4°C in the dark. After two washes, labeled cells were subjected to flow cytometry on a FACScan flow cytometer (Becton Dickinson) and analyzed using the CELLQuest software (Becton Dickinson). Internalization of surface CCR-3 was assayed by incubating cells at 37°C for the indicated lengths of time with 0 -1000 ng/ml human eotaxin or human RANTES (Peprotech, Rocky Hill, NJ). In other experiments, eosinophils were exposed to pertussis toxin (List Laboratories, Campbell, CA) at a dose of 20 -1000 ng/ml for 3 h, and chemokine was added for the last hour of the incubation period. In other experiments, eosinophils were exposed to 1-100 ng/ml staurosporine (Sigma) for 3 h, and chemokine or phorbol 12-myristate 13acetate (PMA; Sigma) was added for the last hour of the 3-h incubation period. Following chemokine exposure, cells were immediately placed on ice and washed with at least twice the volume of cold FACS buffer. Receptor density (percentage) was calculated as 100 ϫ (mean channel fluorescence of chemokine Ϫ mean channel fluorescence of isotypematched control)/(mean channel fluorescence of medium Ϫ mean channel fluorescence of isotype-matched control). Results are expressed as mean Ϯ S.E.
Western Blotting-Whole cell lysates were prepared from eosinophils by washing twice in cold PBS and lysing in radioimmune precipitation assay buffer (1% Nonidet P-40, 0.5% sodium deoxycholate, and 0.1% SDS in PBS) with 10 g/ml aprotinin, 10 g/ml antipain, 10 g/ml chymostatin, 10 g/ml leupeptin, 10 g/ml pepstatin A (all from Boehringer Mannheim) and 2 mM phenylmethylsulfonyl fluoride (Sigma). Detergent-insoluble material was removed by centrifugation at 12,000 ϫ g for 15 min at 4°C. Supernatants were stored in siliconized tubes and either used immediately or stored at Ϫ80°C. The protein concentration was determined using the bicincholic acid assay (Pierce), and 50 g were separated by electrophoresis on a 10% SDS-polyacrylamide gel and transferred to a nitrocellulose membrane. Equal loading was verified by staining with Ponceau S (Sigma). After blocking the membrane for 1 h at room temperature in Tris-buffered saline with 0.2% Tween 20 (TBST) with 5% dry milk, the anti-CCR-3 polyclonal rabbit antiserum (kindly provided by Dr. Bruce Daugherty, Merck) was added for 1 h at room temperature (1:5000 in TBST), followed by goat anti-rabbit horseradish peroxidase-conjugated secondary antibody (1: 10,000 in TBST; Calbiochem). The signal was developed using enhanced chemiluminescence (Amersham Pharmacia Biotech) according to the manufacturer's instructions. Antibody specificity was determined on lysates obtained from HOS.CD4 cells transfected with CCR-1 or CCR-3 (AIDS Research and Reference Reagent Program, Rockville, MD) (27,28). Cycloheximide (Sigma) was used at 10 g/ml for 3 h. Inhibition of protein synthesis (Ͼ80%) was verified by [ 35 S]methionine incorporation for 3 h in the presence or absence of cycloheximide. Protein was precipitated by trichloroacetic acid, and radioactivity measured in a ␤-counter.
Confocal Microscopy-Induced AML cells were cultured in six-well plates on glass coverslips. For experiments, the chemokine was added to the growth medium for 3 h at 37°C. To stop the reaction, cells were placed on ice and fixed with 3% paraformaldehyde in PBS. The fixed cells were washed with PBS, quenched with 15 mM glycine in PBS, and permeabilized with 0.2% saponin in permeabilization buffer (1% cold fish gelatin and 1% bovine serum albumin in PBS). Staining was achieved with the 7B11 antibody (0.75 g/coverslip) in 1% cold fish gelatin and 3% bovine serum albumin in PBS for 1 h at room temperature. Following three washes in 1% cold fish gelatin in PBS, Texas Red-conjugated anti-mouse IgG (Jackson Immunoresearch Laboratories, Inc., West Grove, PA) was added for 1 h at room temperature. Finally, cells were stained with fluorescein isothiocyanate-conjugated anti-CD71 (transferrin receptor) antibody (Research Diagnostics, Inc., Flanders, NJ). Cells were washed three times with 1% cold fish gelatin, twice with PBS, and once with water. Coverslips were mounted onto slides, sealed, and stored at Ϫ20°C until analysis on a Leica DMIRBE inverted microscope equipped with a confocal laser scanner. Images were analyzed with Metamorph (Universal Imaging Corporation, West Chester, PA) and printed in Adobe Photoshop (Adobe Systems Inc., Mountain View, CA). In some experiments, fresh human eosinophils were treated with chemokine for 15 min, cytocentrifuged, and stained for CCR-3 as above.

RESULTS
Ligand-induced Internalization of CCR-3-In order to investigate whether CCR-3 attenuation involves modulation of receptor expression, we examined receptor internalization following ligand binding. First, the surface expression of CCR-3 over 18 h following exposure of eosinophils to eotaxin was investigated by FACS analysis. Eotaxin (100 ng/ml) caused receptor loss, which was detectable after 15 min, remained reduced at 3 h, and returned to base line at 18 h. The combined results are shown in Fig. 1, which demonstrate only 43 Ϯ 9, 43 Ϯ 2, and 76 Ϯ 4% of the original receptor level present on the surface after 15 min, 1 h, and 3 h, respectively. A representative experiment is shown in Fig. 2. Exposure of eosinophils to RAN-TES (100 ng/ml), another CCR-3 ligand, also internalized CCR-3. Only 29 Ϯ 6, 24 Ϯ 2, 24 Ϯ 6, and 47 Ϯ 7% of the receptor remain on the cell surface after 15 min, 1 h, 3 h, and 18 h, respectively (Fig. 1). A representative experiment for RANTES is shown in Fig. 2.
In all instances, RANTES induced a greater magnitude of CCR-3 internalization and had a longer duration compared with eotaxin (p Ͻ 0.05). We were interested in determining if this difference could be related to different potencies of these two chemokines. Therefore, several doses of both chemokines were analyzed. As shown in Fig. 3, RANTES and eotaxin induced a dose-dependent internalization of CCR-3. At all doses, eotaxin was less potent than RANTES. At the highest dose of eotaxin (1000 ng/ml), there was still reduction in CCR-3 surface expression at 18 h. At 100 -250 ng/ml RANTES, there was also reduction in CCR-3 surface expression at 18 h.
We were interested in determining the effect of chemokine removal on the surface expression of CCR-3 following receptor internalization. Eosinophils were exposed to eotaxin or RAN-TES for 15 min, washed extensively to remove chemokine, and then monitored for CCR-3 expression over the subsequent 2 h. After exposure to chemokine, CCR-3 expression was 49, 43, 47, and 88% for eotaxin and 37, 33, 42, and 86% for RANTES at 0, 30, 60, and 120 min, respectively (n ϭ 2-3 for each time point).
Since detection of the receptor was performed using an antibody that might be blocked by ligand binding, we were interested in verifying that ligand binding alone was not sufficient for detection of reduced CCR-3 expression. We therefore performed FACS analysis following chemokine exposure at 4°C. This temperature allows ligand binding but prevents receptor internalization (29,30). At 4°C, there was no internalization of surface CCR-3, whereas replicate cells at 37°C demonstrated receptor internalization (n ϭ 2, data not shown). To demonstrate that this phenomenon was specific for CCR-3, the expression of CD18, an unrelated cell surface molecule, was monitored after treatment with eotaxin (100 ng/ml for 3 h). The CD18 expression level did not change with eotaxin treatment (n ϭ 3, data not shown).
Analysis of CCR-3 Internalization by Confocal Microscopy-In order to directly demonstrate CCR-3 ligand-induced receptor internalization, confocal microscopy was performed.
In these experiments, an immature eosinophilic myelocytic cell line was induced to express CCR-3 by differentiation with IL-5 and butyric acid. Using methods established for differentiation of HL-60 cells, these AML cells express levels of CCR-3 comparable with human eosinophils. 2  cium flux and chemotaxis assays. These cells, unlike fresh eosinophils, express the transferrin receptor, which was useful for co-localization studies. Induced eosinophilic cells were grown on glass coverslips and treated with eotaxin or medium alone for 3 h at 37°C. This treatment resulted in 36 Ϯ 6% (n ϭ 12) CCR-3 expression. Subsequently, the cells were fixed, permeabilized, and stained with the anti-CCR-3 antibody. Nine sections at 0.8 m were taken through each cell. In cells cultured in medium alone, the majority of the CCR-3 staining was at the cell surface (Fig. 4A). Upon exposure to eotaxin, staining with the anti-CCR-3 resulted in a different pattern; the majority of the staining was associated with a cytoplasmic granular staining (Fig. 4B). Experiments were conducted to further delineate CCR-3 localization. Transferrin receptor is a well established marker of early endosomes (31). Two-color staining was used to identify CCR-3 with Texas Red (depicted in red) and the transferrin receptor with fluorescein isothiocyanate (depicted in green). Co-localization of both markers produces a yellow color. Untreated cells had co-localization of both markers predominantly on the membrane (Fig. 4C). In contrast, following eotaxin treatment, CCR-3 and the transferrin receptor co-localized to a predominantly intracellular location (Fig.  4D). Eotaxin-treated cells did not stain for CCR-3 alone, suggesting that the major intracellular location overlapped with that used by the transferrin receptor. Experiments were also conducted with fresh human eosinophils in order to verify that ligand-induced receptor internalization, rather than antigen blockade by chemokine, was occurring. These experiments revealed translocation of CCR-3 staining predominantly from a membrane-associated pattern to a granular pattern in a perinuclear location consistent with endosomes (data not shown).
Ligand-induced Modulation of CCR-3 Protein Level-Inter-nalization of CCR-3 may induce receptor degradation. Therefore, it was of interest to examine total cell protein for the level of CCR-3 protein by Western blot analysis. The conditions for Western blot analysis were first established from control cells: human eosinophils, untransfected HOS.CD4 cells, or HOS.CD4 cells transfected with CCR-1 or CCR-3. A polyclonal rabbit anti-human CCR-3 serum detected multiple background bands but showed a strong, large band between 50 and 60 kDa in HOS.CD4 cells transfected with CCR-3 but not with CCR-1 (Fig. 5A). Human eosinophils had a strong band between 50 -60 kDa, and this band was not present in granulocytes or mononuclear cells (Fig. 5B). In all cases, detection of immunoreactive CCR-3 protein was lost if the protein extract was boiled prior to electrophoresis. Eosinophils were then treated with medium alone or medium supplemented with eotaxin (250 ng/ml) or RANTES (100 ng/ml) for 3 h, conditions that promote optimal chemokine-induced internalization. Western blot analysis of whole cell lysates revealed that the level of CCR-3 protein decreased by ϳ30% (Fig. 5C). The decreased level of CCR-3 was further reduced (by ϳ60%) when eosinophils were treated with cycloheximide for the 3 h during the exposure to the chemokine. Cells treated with cycloheximide alone had no change in the level of immunoreactive CCR-3 protein compared with cells incubated in medium alone (Fig. 5C). Since the antibody used in the Western blot recognizes the carboxylterminal region of CCR-3, the absence of degradation products may be due to the loss of this epitope in the degradation products. These experiments demonstrate that internalization of CCR-3 is accompanied by receptor degradation and that de novo synthesis of CCR-3 protein significantly contributes to the total level of CCR-3 protein following chemokine binding. In order to investigate if reexpression of CCR-3 on the cell surface  A and B). Alternatively, cells were co-stained with the anti-transferrin receptor antibody anti-CCR-3 (C and D). Staining was analyzed using confocal microscopy. CCR-3 staining is depicted in red, and transferrin receptor staining is shown in green. Co-localization is shown as yellow.
is dependent upon protein synthesis, we monitored CCR-3 expression by FACS analysis on cells treated with eotaxin in the presence or absence of cycloheximide. Following treatment with 100 ng/ml eotaxin for 3 h, the receptor was beginning to reappear on the cell surface. In contrast, on cells treated with eotaxin in the presence of cycloheximide, the receptor expression on the cell surface remained low (Fig. 5D). These data suggest that reexpression of CCR-3 on the surface depends in part upon de novo protein synthesis.
Effect of Pertussis Toxin on CCR-3 Internalization-We were interested in determining if G-protein coupling was required for CCR-3 internalization. CCR-3-induced calcium transients are known to be inhibited by pertussis toxin, suggesting that CCR-3 couples to G i -proteins (32). We first determined a dose of pertussis toxin that was able to completely inhibit eotaxininduced calcium transients in eosinophils. Treatment of eosinophils with 100 -1000 ng/ml of pertussis toxin for 3 h inhibited calcium flux, but we were surprised to see that these doses also reduced CCR-3 expression down to 30% (data not shown). At a lower dose of 20 ng/ml pertussis toxin, inhibition of eotaxininduced calcium transients was maintained (Fig. 6, A and B), but the level of CCR-3 was not significantly reduced. Exposure of eosinophils to this dose of pertussis toxin (20 ng/ml) did not block eotaxin (100 ng/ml)-induced receptor internalization (Fig. 6, C and D). There was also no evidence of an effect of pertussis toxin using eotaxin at 10 or 500 ng/ml (data not shown). These data indicate that CCR-3 internalization is not dependent upon G i -protein coupling or calcium transients in human eosinophils.
Involvement of Protein Kinase C in CCR-3 Internalization-We were next interested in determining if protein kinase C (PKC) was involved in ligand-induced CCR-3 internalization, since the internalization of some GPCRs is dependent upon PKC. We first treated human eosinophils with PMA in order to determine if pharmacological activation of PKC was able to cause CCR-3 down-regulation. As shown in Fig. 7A, treatment of eosinophils with PMA for 1 h resulted in a dose-dependent down-modulation of CCR-3 surface expression. We next determined the nontoxic dose of staurosporine that would effectively inhibit PMA-induced CCR-3 down-regulation. As shown in Fig.  7B, pretreatment of eosinophils with 10 ng/ml of staurosporine for 2 h prior to PMA completely inhibited PMA induced CCR-3 down-modulation. We next determined the effect of staurosporine treatment on chemokine-induced CCR-3 internalization. As shown in Fig. 7C, staurosporine had no effect on the level of eotaxin (100 ng/ml)-induced CCR-3 internalization. There was also no evidence of an effect of staurosporine using eotaxin at 10 or 500 ng/ml (data not shown). These results indicate that although PMA induces CCR-3 down-modulation, PKC is not involved in ligand-induced CCR-3 internalization. DISCUSSION We report that CCR-3 is rapidly removed from the cell surface into an intracellular endocytic compartment following ligand binding. The two CCR-3 ligands studied, eotaxin and RANTES, induced a different magnitude and duration of receptor internalization. In all cases, RANTES was more potent and induced a longer duration of internalization than eotaxin although it was not a stronger chemoattractant in previous studies (15) and in eosinophil transmigration assays (data not shown). This suggests that internalization of CCR-3 involves molecular events that are dissociated, at least in part, from the signals involved in triggering chemotaxis. Consistent with this, chemokine-induced internalization was not dependent upon G-protein coupling or calcium transients although both of these processes are involved in chemoattraction. Interestingly, a derivative of RANTES that functions as an antagonist, aminooxypentane-RANTES, induces a stronger and more prolonged internalization of CCR-5 in transfected Chinese hamster ovary cells compared with RANTES itself (33). Additionally, IL-8 derivatives that have equal chemotactic activity can have differential ability to induce elastase release, suggesting that they trigger independent events involving the same receptor (34).
We also demonstrate that the half-life of CCR-3 is prolonged (Ͼ3 h), since treatment of eosinophils with cycloheximide did not change the level of CCR-3 protein in the whole cell lysate. However, following 3 h of chemokine treatment, the level of cell-associated CCR-3 protein was decreased in a manner augmented by cycloheximide treatment. This indicates that chemokine treatment induces protein degradation and that new protein synthesis is involved in maintaining the level of the CCR-3 protein following chemokine treatment. Concomitantly, the surface expression of CCR-3 was further decreased with cycloheximide treatment, indicating that reexpression of the receptor on the cell surface is dependent upon de novo protein synthesis. Ligand-induced modulation of chemokine receptor expression has only been examined for a limited number of chemokine receptors, and variable pathways have been reported. IL-8 induces rapid internalization of its receptor in neutrophils (29). Stromal cell-derived factor-1␣ induces a decrease of cell surface CXCR-4 in the CEM T-cell line, HeLa cells, and peripheral blood mononuclear cells (35,36). Additionally, CCR-5 ligands induce receptor internalization in lymphocytes, monocytes/macrophages, and CCR-5-transfected Chinese hamster ovary cells (33). In these cases, the receptors enter an endocytic pathway but recycle by 3 h after ligand binding. In contrast, CXCR-2 undergoes internalization and does not recycle, since it enters a degradative lysosomal pathway (37). Rapid ligand-induced internalization of CCR-1 in transfected Chinese hamster ovary cells (38) and CCR-2b in transfected HEK-293 cells (39) have been described, but their intracellular processing was not studied. The diverse mechanisms of ligand-induced modulation of chemokine receptors indicate the importance of dissecting these processes for each chemokine receptor.
Because most experiments were done by FACS analysis utilizing an antibody against CCR-3, it remained possible that the apparent decrease in the level of CCR-3 expression was dependent upon blockade of the CCR-3 immunoreactive epitope by the chemokine. Therefore, experiments were conducted at 4°C in order to block internalization. These experiments showed no modulation of CCR-3 expression, indicating that ligand binding alone was not sufficient to decrease CCR-3 detection. Furthermore, confocal microscopy was performed in order to directly demonstrate ligand-induced CCR-3 internalization. These experiments were performed on an eosinophilic cell line that expressed high levels of functionally active CCR-3. This cell line was useful especially because it maintained expression of the transferrin receptor, which allowed co-localization studies to be performed. On resting cells, most CCR-3 was demonstrated to be on the cell membrane; however, following ligand treatment, the majority of the CCR-3 protein was detected in an intracellular location. Co-staining studies with the transferrin receptor revealed that intracellular CCR-3 was predominantly co-localized with the transferrin receptor, which is known to recycle via the early endosome compartment. This . Replicate cells were also exposed to eotaxin for the last 1 h of the culture, and the level of CCR-3 expression was determined (C, D). Cell surface expression of CCR-3 was measured by FACS analysis and compared between eotaxin-treated cells (dashed histogram) and non-chemokine-treated cells (boldface histogram). The level of isotype matched control antibody expression is indicated with the filled histogram. The insets in C and D represent data expressed as percentage of CCR-3 expression. pathway exists in numerous cells and is commonly involved in receptor endocytosis. It is interesting that the cytoplasmic motif (YXXF) required for internalization of other receptors (e.g. transferrin receptor) is conserved in CCR-3 (40).
We have begun to elucidate mechanisms of CCR-3 internalization. Experiments conducted with pertussis toxin demonstrated that G i -protein coupling is not necessary for internalization. Interestingly, high doses of pertussis toxin alone downregulated the level of surface CCR-3 expression. The dissociation of G-protein coupling and GPCR internalization has been reported for other selected GPCRs such as CXCR-4 (35,36). We also tested the involvement of PKC in mediating ligand-induced CCR-3 internalization. Using staurosporine to inhibit PKC, we demonstrated that ligand-induced CCR-3 internalization was not dependent upon PKC. However, in these experiments, CCR-3 down-modulation was induced by activation of PKC with PMA. Similarly, although PMA and ligands induce rapid phosphorylation of CXCR-2, only PMA-induced phosphorylation is inhibited by staurosporine (37). These data indicate that at least two pathways exist for down-modulating CCR-3 expression: one mediated by PKC and another mediated by chemokines and independent of PKC. The latter pathway may be dependent upon G-protein-related kinases, such as GRK-2, which has been shown to be involved in CCR-5 internalization (41).
The present results have several biological implications, especially since CCR-3 events have been investigated specifically in eosinophils. Conclusions concerning eosinophil chemokine receptor events have often been drawn from investigation of heterologous cell lines transfected with CCR-3. Since individual cell types often utilize distinct signaling events, it is important to examine chemokine-triggered biochemical events in eosinophils. For example, fMLP inhibits adenylate cyclase in fMLP receptor-transfected cells but not in human neutrophils (25). We have therefore performed all experiments in eosinophilic cells. The finding that chemokines induce CCR-3 to undergo significant and prolonged receptor internalization has mechanistic implications for understanding eosinophil trafficking in vivo. It is widely accepted that chemokines induce cellular activation and chemoattraction. However, it is unclear if chemokines are also involved in stopping leukocyte movement. For example, eotaxin has been demonstrated to be required for the maintenance of gastrointestinal eosinophils at base line (23). If eotaxin also induces receptor internalization with subsequent cellular hyporesponsiveness, then eosinophils would likely home into the intestine and become localized there, since they would no longer be responsive to other chemokine gradients operating through CCR-3. In contrast, eosinophils in hematopoietic organs, tissues that do not express eotaxin, would remain responsive to the induction of subsequent chemokine gradients. Additionally, the observed inefficiency of receptor recycling following ligand binding, especially for RANTES, may have mechanistic implications concerning the difficulty in expressing CCR-3 on the surface of transfected cell lines. In one study, only 2-5% of CCR-3-transfected cells expressed CCR-3 on their surface although substantial intracellular protein could be detected in most transfected cells (15), suggesting a problem with CCR-3 protein trafficking. Last, the current study sheds light on possible mechanisms by which eotaxin may block cellular entry of CCR-3 trophic human immunodeficiency virus type 1 strains (13). Besides directly blocking human immunodeficiency virus binding to CCR-3, eotaxin may inhibit human immunodeficiency virus uptake by causing internalization of the CCR-3 co-receptor. Further studies are under way analyzing the long term biological consequences of exposing eosinophils to chemokines and a further elucidation of the biochemical events associated with CCR-3 signaling. tained through the AIDS Research and Reference Reagent Program, Division of AIDS, NIAID, National Institutes of Health, from Dr. Nathaniel Landau. RANTES and macrophage inflammatory protein-1␣ were also obtained through the AIDS Research and Reference Reagent Program, Division of AIDS, NIAID, National Institutes of Health, from Peprotech.